Aharrypotter commented on PR #20114: URL: https://github.com/apache/tvm/pull/20114#issuecomment-5269916228
> Overall this looks good, but there are a couple of missing edge cases around `COMPLEX64`. Relax represents complex tensors with an extra trailing size-2 axis, so reshape needs to use `_get_relax_tensor_shape`, and transpose needs to append that pair axis to the permutation. Using the raw TFLite output shape also breaks rank-0 reshape outputs. Could we handle those cases and add scalar/complex tests? Done. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
